Autocodificadores Variacionales (VAE) Fundamentos Teóricos y Aplicaciones
It provides an introductory resource for Spanish-speaking researchers and practitioners, but is incremental as it reviews existing VAE fundamentals without new contributions.
The paper introduces Variational Autoencoders (VAEs) as probabilistic graphical models that encode input data into a latent space and reconstruct it, enabling generation of new data similar to the original distribution. It is written in Spanish to make this knowledge accessible to Spanish-speaking audiences.
VAEs are probabilistic graphical models based on neural networks that allow the coding of input data in a latent space formed by simpler probability distributions and the reconstruction, based on such latent variables, of the source data. After training, the reconstruction network, called decoder, is capable of generating new elements belonging to a close distribution, ideally equal to the original one. This article has been written in Spanish to facilitate the arrival of this scientific knowledge to the Spanish-speaking community.